Identify optimal cluster resolution of scale-free shared nearest neighbor network (SNN)

Identify optimal cluster resolution of scale-free shared nearest neighbor network (SNN)

SSNResolution(
  object,
  graph,
  target.purity = 0.7,
  start.res = 0.5,
  step.size = 0.05,
  target.level = "global",
  max.iter = 100,
  verbose = F
)

Arguments

object

Seurat object

graph

name of SNN graph in `object`.

target.purity

Target purity to find resolution for.

start.res

Initial resolution to begin optimization with. Default is 0.5.

step.size

Step size between consecutive resolutions to test. Default is 0.05.

target.level

Specify whether `target.purity` is achieved at "global" level [i.e., median(purity)] or at "cluster" level. Default is "global".

max.iter

Maximum number of iterations. If `max.iter` is reached, non-optimal resolution may be returned.

verbose

Show progress. Default is T.

Value

optimal cluster resolution

References

https://nmikolajewicz.github.io/scMiko/articles/Module_Detection.html

See also

neighborPurity

Author

Nicholas Mikolajewicz